ACCESS ExperimentalandModelingStudiesofSr andCs SorptiononCryogelsandComparisontoCommercialAdsorbents

Experimental and Modeling Studies of Sr

²⁺

and Cs

⁺

Sorption on Cryogels and Comparison to Commercial Adsorbents

Alzhan Baimenov, Fabio Montagnaro, Vassilis J. Inglezakis,* and Marco Balsamo*

Cite This:Ind. Eng. Chem. Res.2022, 61, 8204−8219 Read Online

ACCESS

Metrics & More Article Recommendations

*

^sı Supporting Information

ABSTRACT: In this work, two cryogels with the key monomers methacrylic acid and 2-acrylamido-2-methyl-1-propansulfonic acid (named AAC and SAC, respectively) with various functional groups were used as adsorbents for the removal of cesium and strontium ions from aqueous solutions. Kinetics, equilibrium, and column studies were carried out including experiments in different water matrices (ultrapure, tap, and river water) and comparison to commercial adsorbents. AAC reached sorption capacity of 362 mg g⁻¹for Cs⁺and 209 mg g⁻¹for Sr²⁺, whereas SAC polymer showed maximum removal capacities of 259 and 211 mg g⁻¹for Cs⁺ and Sr²⁺, respectively. The five cycles of adsorption/desorption experiments showed a maximum of 8% loss of effectiveness for

both cryogels. Batch kinetics adsorption data were modeled by using a rigorous diffusional model coupled to a novel fractal-like expression for variable surface diffusivity. The model revealed that the surface diffusivity dependence on time is nonmonotonic, with the occurrence of a maximum. Also, bothfluid film and intraparticle transport resistances were shown to be important, with the internal one being more influential. The cryogels and two commercial materials (ion-exchange resin and zeolite) were tested for the removal of Cs⁺ and Sr²⁺in ultrapure, tap, and river water; the results showed that the cryogels exhibit competitive effectiveness.

1. INTRODUCTION

With the growth in energy consumption, the development of alternative energy sources is becoming increasingly popular.

Despite several advantages of nuclear power, its use is not risk- free, and accidents can be detrimental to people and the environment such as what happened in Chernobyl and Fukushima.¹ Among the harmful radionuclides leaked from nuclear power plants accidents, radioactive cesium (¹³⁷Cs) and strontium (⁹⁰Sr) have substantial toxic effects, emitting beta- particles andγ-rays. Moreover, due to the chemical similarity of ¹³⁷Cs and ⁹⁰Sr to K and Ca, respectively, the isotopes released after a nuclear accident can be easily incorporated into terrestrial and aquatic organisms. Both Cs and Sr radioisotopes have half-lives around 30 years; therefore, it is imperative to immediately and effectively remove them from wastewater and contaminated water bodies.²

Removal of radioactive ions from wastewater has been the focus of numerous studies.²⁻⁶ Several methods have been considered for the recovery of metal ions from wastewater including chemical precipitation,⁷ chemical coagulation/

flocculation,⁸ membrane filtration,⁹ ion-exchange processes,¹⁰ bioremediation,¹¹and adsorption.¹²Typical adsorbents such as activated carbons (AC), natural zeolites, and clays have been used, but they are often characterized by low adsorption capacities or the need to be chemically treated to acquire desirable adsorption properties.¹³⁻¹⁵ Natural sorbents like zeolites and AC usually show low removal capacities in the

ranges of 1.3−10 mg/g for Sr²⁺ ions and 0.7−51 mg g⁻¹for Cs⁺. Synthetic polymers like hydrogels and cryogels can reach much higher removal capacities, up to 312 mg g⁻¹for Sr²⁺and 420 mg g⁻¹ for Cs⁺ ions.^1,4,16−33 A list of literature works investigating Cs⁺and Sr²⁺ions adsorption from aqueous media onto different sorbents is reported inTable S1.

Certain properties are desirable when it comes to an effective cation adsorbent, e.g., (i) the presence of chelating sites or ion-exchange groups, (ii) a hydrophilic and three- dimensional (3D) network structure to uphold water-flow for an extended period of time, (iii) cost-effectiveness, (iv) facile synthesis, and (v) easily regeneration. Natural materials, like zeolites or charcoals, are low-cost, but without additional modification, they are nonselective and may have low removal efficiencies. Among several synthetic materials, cryogels are highly effective sorbents toward heavy metals ions.^34,35Upon appropriate modification, the functional groups in their network structures, such as carboxyl (−COOH), hydroxyl (−OH), amine (−NH2), and/or thiol (−SH), can provide the

Received: February 15, 2022 Revised: May 4, 2022 Accepted: May 19, 2022 Published: June 2, 2022

© 2022 The Authors. Published by

Downloaded via 85.117.116.93 on February 13, 2023 at 07:30:49 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

necessary chelating sites for effective cation removal from aqueous solutions. Further advantages include the ease of regeneration and biodegradation. Despite the large number of published studies investigating adsorption kinetics modeling for different adsorbent/adsorbate systems, most use empirical expressions, e.g., pseudo-first- or pseudo-second-order models, not suitable for studying the underlying mechanisms and mass transfer phenomena. Some studies apply simplified diffusion- based models not mirroring the complex mechanisms taking place during surface diffusion, for instance, the occurrence of nonmonotonic patterns of the surface diffusivity in the course of the sorption process.^36,37 In the case of adsorption onto cryogels, a limited number of papers apply rigorous dynamic models; for example, a rate constants distribution model also accounting for diffusion limitations was recently proposed to interpret Cu(II) adsorption on a polyethylenimine cryogel.³⁸ In the present paper, a detailed understanding of intraparticle transport mechanisms is gained by applying a diffusion-based model accounting forfluidfilm, pore, and surface transport of the adsorbate. In particular, the element of novelty in dynamic modeling relies in adopting our recently proposed fractal-based expression for surface diffusivity able to depict the above- mentioned nonmonotonic patterns in the time domain.³⁹

In order to evaluate the impact of different functional groups on the removal of Cs⁺ and Sr²⁺, two different cryogels were synthesized by free-radical copolymerization of acrylate-based precursors with allylamine under subzero conditions. Thefirst cryogel (AAC) is composed of copolymers of methacrylic acid, allylamine, dimethylacrylamide, and methylenbis(acrylamide).

The second cryogel is allylamine-2-acrylamido-2-methyl-1- propansulfonic acid (SAC) and has a chemical composition of monomers and concentrations similar to those of AAC, but methacrylic acid is here substituted by 2-acrylamido-2-methyl- 1-propansulfonic acid. The cryogels were fully characterized and applied for the removal of Sr²⁺and Cs⁺ions from aqueous solutions. Kinetics and equilibrium studies were conducted, and dynamic adsorption results were interpreted by application of rigorous diffusion models including variable surface diffusivity expressions, a topic rarely investigated in the pertinent literature. The adopted fractal-like expression for the surface diffusivity allowed us to study the change of diffusivity during the adsorption process and quantitatively evaluate the relative importance of intraparticle versus fluid film mass transfer resistance. In combination with postsorption characterizations and obtained adsorption results, potential removal mechanisms were proposed. The effect of different water matrices and the comparison to commercial adsorbents were also studied. To the best of our knowledge, this work represents the most detailed experimental and modeling investigation for the removal of Sr²⁺ and Cs⁺ ions by use of cryogels.

2. MATERIALS AND METHODS

2.1. Reagents. For the synthesis of cryogels, N,N- dimethylacrylamide (DMAAm, 99%), allylamine (AA) (98%), 2-acrylamido-2-methyl-1-propanesulfonic acid (AMPS), methacrylic acid (MAAc, 99%), N,N-methylenebis (acrylamide) (BisAAm, 99%), 70% H₃PO₄, 5 M NaOH, ammonium peroxodisulfate (APS, 98%), and N,N,N′,N′- tetramethyl ethylene diamine (TEMED, ≥99.5%) were used.

Sr(NO₃)₂ and CsNO₃ (purity of 99%) salts were used as strontium and cesium sources for all adsorption experiments.

All reagents were purchased from Sigma-Aldrich (Germany).

Ultrapure water (UPW) purified by a Puris MR-RO1600 (Mirae ST, South Korea) reverse osmosis apparatus was used for the preparation of cryogels and metals solutions.

2.2. Synthesis of AAC and SAC Cryogels.Two cryogels with different compositions, named AAC and SAC, were synthesized at subzero temperatures by free-radical polymer- ization according to previously reported procedures.⁴⁰⁻⁴² Briefly, MAAc or AMPS for AAC and SAC cryogels, respectively, and the cross-linking agent BisAAm were mixed in 10 mL of preliminary degassed UPW and neutralized by 5 M NaOH. In another beaker, DMAAm and AA were added to 7.5 mL of UPW and acidified with concentrated H₃PO₄under continuous stirring. Afterward, both solutions were inter- mingled and TEMED was added dropwise, followed by vigorous stirring and cooling in an ice bath for 30 min under N₂ atmosphere. After the cooling and degassing stages, 0.25 mL of 5 wt % of APS was added. Finally, 2 mL of the monomeric mixture was poured into plastic syringes of 1 cm diameter. Then, sealed syringes were placed in an ethanol cooled cryobath (Julabo F34, Germany) and kept at−12°C for 24 h. The resulted monolithic cryogels were thawed out at room temperature and washedfirst with 10% ethanol and then with 2 L of UPW to remove unreacted reagents. For the further use, both types of cryogels were lyophilized on a FreeZone 2.5 L (Labconco, Kansas City, MO) instrument at

−45 °C and under vacuum (0.4 mbar) for 48 h, in order to remove water.

2.3. Adsorbents Characterization. The morphological structure of cryogels coated by 7 nm Au was evaluated by using a Zeiss Crossbeam 540 scanning electron microscope (SEM) at 20 kV. Semiquantitative energy-dispersive X-ray (EDX) spectrometer (INCA X-sight, Oxford Instruments) connected to the SEM was used for characterizing point and surface elemental analysis before and after metal ions removal. For the determination of the cryogels structure byfluorescent confocal laser scanning microscope LSM 780 (Zeiss, Germany), the polymer discs were immersed in 0.5 M solution of rhodamine B for 24 h and then washed by water to remove dye from the polymer pores. Nitrogen adsorption/desorption experiments at

−196°C were carried out in an Autosorb IQ (Quantachrome Instruments, Ashland, VA) porosimeter to determine the cryogels textural properties. In particular, the micromesopo- rous surface area was evaluated according to Brunauer− Emmett−Teller (BET), Barrett−Joyner−Halenda (BJH), and Dubinin−Radushkevich (DR) models applied to raw adsorp- tion data. Prior to the porosimetric analysis, lyophilized cryogel samples were degassed at 105°C for 6 h to remove oxygen and humidity in pores. The X-ray photoelectron spectroscopy (XPS) measurements were conducted on a VG-Microtech Multilab 3000 device equipped with a 9-channeltrons hemi- spherical electron analyzer and X-ray radiation source with Mg and Al anodes. The binding energies (BE) were calibrated by a C 1s core level at 284.8 eV as a reference.

2.4. Adsorption Studies. 2.4.1. Effect of pH. To assess the effect of pH on the adsorption efficiency of cryogels, a series of experiments was carried out in the pH range 1.0−5.0 at room temperature using 0.07 g of polymers in 100 mL of metal ion solutions with concentrations of 100 mg L⁻¹. The pH of Sr²⁺and Cs⁺solutions with initial concentration of 100 mg L⁻¹ was adjusted with HNO₃ to achieve the desired pH values. The residual concentrations of metal ions were analyzed by iCAP RQ ICP-MS analyzer (ThermoFisher Scientific, Waltham, MA).

The amount of adsorbed metal was calculated by material balance:

= −

q C C V

m

( )

eq

0 eq

(1) The reader is referred to the “Nomenclature” section for the meaning of the symbols.

2.4.2. Batch Adsorption Kinetics. The Sr²⁺ and Cs⁺ solutions were prepared by dissolving analytical-grade Sr- (NO₃)₂ and CsNO₃ in UPW, respectively. A volume of 100 mL of a solution of Sr²⁺and Cs⁺at 100 mg L⁻¹concentration was mixed without pH regulation (pH 5.5) with 0.07 g of cryogel in plastic tubes with shaking at 120 rpm at room temperature. At predetermined time intervals, a 0.1 mL sample volume was taken from the tubes and analyzed using ICP-MS (iCAP RQ, ThermoFisher Scientific, Waltham, MA) with the appropriate dilution. The total sampling volume in all experiments was kept below 3% of the original volume. Blanks with the same initial metal ion concentration and volume, without solids, were analyzed. Blank experiments showed that the loss of Sr²⁺ and Cs⁺ due to adsorption on the tube walls was less than 2%. All experiments were carried out in duplicate;

the experimental standard deviation was less than 5%. The specific amount of Sr²⁺and Cs⁺adsorbed at each process time was calculated as

= [ − ]

q t C C t V

( ) ₀ m( )

(2) Batch kinetics data of Sr²⁺and Cs⁺adsorption onto cryogels were modeled according to a rigorous diffusional model accounting for both pore volume and surface transport mechanisms, namely the pore volume surface diffusion model (PVSDM) as presented by Ocampo-Pérez et al.⁴³The assumptions and equations of the PVSDM model are presented in theSupporting Information (eqs S1−S5).

Despite Langmuir and Freundlich equilibrium models allowing a general good interpretation of the whole adsorption isotherms (including equilibrium points at concentrations higher than those used for the batch kinetic run), a poorfitting of dynamic data for longer process times occurred when applying them as local isotherms to analyze the batch dynamic data. The equilibrium loading at low concentrations was generally underestimated by the above-mentioned models, thus resulting in unreliable dynamic transport parameters. As a consequence, the Henry model was selected as more adequate to describe local equilibrium at low concentrations, i.e., dilute liquid phase for the kinetics run modeling:

=

q t r( , ) K C t r_{H p}( , ) ₍₃₎ The PVSDM was applied to analyze kinetic data by considering both a constant surface diffusivity D_s and expressions accounting for its variation along the adsorption

process. In particular, the Chen and Yang’s expression,⁴⁴here abbreviated as CY (eq 4), and our recently proposed fractal- based equation, abbreviated as MBI in Inglezakis et al.³⁹(eq 5), based on our previous works on the topic, e.g., Montagnaro and Balsamo⁴⁵and Balsamo and Montagnaro,⁴⁶were adopted:

θ

θ θ θ λ λ θ

θ θ

= − + − + [ − ] −

− +

λ λ

(

λ

)

D D H

( ) 1 (2 ) 1 (1 )

1

s s0 2 2

2

2 2

(4)

β β

= + + −

+ ^α

−

Ä ÇÅÅÅÅÅ ÅÅÅÅÅ

É ÖÑÑÑÑÑ ÑÑÑÑÑ

D t D t t

1 ( )

1

( 1)

(1 )

( 1)

s s0 h (5)

Both equations are characterized by a robust theoretical basis and, with respect to other expressions commonly found in the literature, they allow to predict increasing, decreasing and nonmonotonic patterns of the surface diffusivity. While the CY expression (eq 4) accounts for a D_s dependence on the surface coverage degree,θ=q(t,r)/q_m, in the MBI expression (eq 5), the surface diffusivity is a function of the process time.

In the following, the versions of the PVSDM including the CY and MBI equations are termed PVSDM-CY and PVSDM-MBI, respectively.

In order to compare the relative importance of intraparticle versusfluidfilm mass transfer resistance, the Biot number can be invoked.⁴⁷In particular, for the analysis of the PVSDM-MBI case, we define in the present work an average Biot number Bi t( )as

= ρ

Bi t k RC+ D t q D C

( ) ( )

f 0

s p 0 p 0 (6)

where D t_s( ) represents the mean integral value of the surface diffusivity function expressed byeq 5.

The solution of the partial and ordinary differential equations constituting the PVSDM, PVSDM-CY, and PVSDM-MBI, was carried out by adopting the method of lines implemented in the software MATLAB R2019b for the discretization of the particle radial coordinate. The bestfitting parameters for the numerical models were determined by means of the lsqnonlin algorithm of MATLAB, which minimizes the sum of squared errors between experimental and theoretical values of pollutant concentration in the bulk solution. In order to reduce the number of fitting parameters for each model, the following numerical strategy was adopted at fixed sorbent/adsorbate system: (i) The k_f parameter was preliminary determined through fitting experimental data by means of the simpler homogeneous surface diffusion model, which accurately predicted adsorption data in particular in the first stages of the process, where fluid film resistance is more relevant. (ii)D_s0wasfixed for both PVSDM-CY and PVSDM- MBI as the bestfittingD_svalue obtained from the application Table 1. Sorbents Textural Properties,Dm, andDp Values Adopted as Input Parameters for the Dynamic Models

textural parameters

sorbent R[m] εp[−] ρp[g m⁻³] adsorbate D_m[m²s⁻¹]^a sorbent/adsorbate D_p[m²s⁻¹]

SAC 4.84×10⁻³ 0.85 4.38×10⁴ Cs⁺ 2.06×10⁻⁹ SAC/Cs⁺ 1.15×10⁻⁹

SAC/Sr²⁺ 8.81×10⁻¹⁰

AAC 5.02×10⁻³ 0.89 3.95×10⁴ Sr²⁺ 1.58×10⁻⁹ AAC/Cs⁺ 1.31×10⁻⁹

AAC/Sr²⁺ 1.01×10⁻⁹

aAfter Kadhim and Gamaj.⁴⁸

of PVSDM. Consequently,D_swas set asfitting parameter for PVSDM, λ for PVSDM-CY and the triad h, α, and β for PVSDM-MBI.Table 1reports the sorbents textural properties, D_m values retrieved from literature for Sr²⁺ and Cs⁺, and calculated D_p values, all adopted as input parameters for the numerical dynamic models. The relationship between D_mand D_pis provided ineq S4.

2.4.3. Equilibrium Adsorption Isotherms. A set of three equilibrium adsorption isotherms was obtained in batch mode without pH adjustment for each investigated adsorbate/

adsorbent system. The difference among experiments is the use of various mass of cryogels and initial concentrations of Sr²⁺ and Cs⁺, as shown in Table 2. All experiments were

conducted in plastic tubes at least in duplicate under shaking, by use of a Rotamax 120 (Heidolph, Germany) at room temperature, by sampling and measuring the residual Sr²⁺and Cs⁺ daily until equilibrium was reached. The total liquid sampling volume was less than 3%. The average relative standard deviation for all experiments was 1.5%, while on graphs the absolute standard deviation values will be reported for each solid-phase loading.

Since analytical errors are inevitable, it is necessary to take into account that if the difference between the initial and equilibrium concentrations of the adsorbate in the aqueous phase is not sufficiently large, then considerable errors and unreliable results may occur when calculating the equilibrium capacity of the sorbent. This is especially important when using a small amount of adsorbent and a large concentration of adsorbate. It was calculated that for a relative analytical concentration error of about 2.5% and a desired capacity

relative error of less than 15%, the equilibrium concentrations of the aqueous phase should be kept below 85% of the initial concentration. To avoid significant uncertainties, this rule was followed in the present study.

The results of the experiment 2 were adopted to evaluate the fitting of experimental data with respect to the widely used Langmuir and Freundlich isotherm models, as expressed byeqs 7and8, respectively:⁴⁹

= +

q q K C

C K

eq 1

m L eq

eq L (7)

=

q_eq K C_{F eq}^1/n ₍₈₎

The Langmuir sorption isotherm assumes the monolayer adsorption of adsorbate onto the homogeneous surfaces, while the Freundlich sorption isotherm corresponds to multilayer adsorption on energetically heterogeneous surface. The total release of Na⁺ ions from both types of cryogels in metal-free solutions was evaluated by placing 80 mg of dry polymer in 100 mL of UPW under shaking at 120 rpm for 30 days. The investigated sodium release was around 1.3 mg g⁻¹for both cryogels.

2.4.4. Leaching Experiments.The release of the metal ions from the cryogels after adsorption was studied by leaching experiments at pH 6.5. Following the equilibrium experiments, the cryogel samples with the highest loading were washed with 1 L of UPW in tightly closed containers, and the systems were left for 30 days under shaking (120 rpm) at room temperature.

The average BJH pore size is considerably smaller than this identified by SEM images, probably due to resolution limitations related to SEM. The samples were withdrawn from each container and analyzed for leached Sr²⁺and Cs⁺. All the leaching experiments were carried out in duplicate, and the average values are reported.

2.4.5. Removal from Various Water Matrices and Comparison to Commercial Adsorbents. Removal studies were conducted in batch mode by mixing 10 mg of each adsorbent with 50 mL of metal solution at 10 mg L⁻¹ Table 2. Parameters of Isotherm Adsorption Experiments

no. of experiment

mass of cryogels [g]

initial concentration [mg L⁻¹]

solution volume [mL]

1 0.0022−0.150 100 100

2 0.07 100−1000 100

3 0.07−0.9 1000 100

Figure 1.Schematic representation of the formation of SAC cryogel.

concentration in different water matrices. UPW, tap water (from Nazarbayev University laboratories), and river water (sampled from 43°15′4.0″N 76°51′50.7″E, Bolshaya Alma- tinka, Almaty, Kazakhstan) were used without any further purification. The efficiency of cryogels were compared with commercial ion exchange materials, namely, an H⁺-form ion- exchange resin (Merck) and NaY synthetic zeolite (Sigma- Aldrich). Sampling was done at 30 min and 2 and 24 h, and the residual Sr²⁺and Cs⁺concentrations were measured by Dionex ICS 6000 ion chromatograph, afterfiltering through 0.45μm hydrophilicfilter and dilution. The experiments were done in duplicate, and the average values are reported.

2.4.6. Column Experiments.A dry cryogel monolith (1 cm height and 0.7 cm diameter) was placed into a plastic column (15 cm height and 0.8 cm internal diameter) and wetted with UPW. Upon contact with water, the cryogels diameter increased by 1 mm, and as a result, the monoliths adhered tightly to the column walls, which excludes the possibility of the passage of aqueous metal solutions between the cryogel and the column walls, i.e., bypass phenomena. A down-flow

rate of 2.5 mL min⁻¹ was maintained by using a peristaltic pump (Gilson, Minipuls 3). The experiments were carried out at room temperature, at 100 mg L⁻¹ of Sr²⁺ and Cs⁺ concentration. Samples were collected from the outflow every minute, and they were analyzed as described in the previous sections. Three simplified models, namely, Thomas’, Yoon−Nelson’s, and Adams−Bohart’s models, were adopted tofit the breakthrough data. The models are presented in the Supporting Information eqs S6−S8.

2.4.7. Regeneration and Reusability of Cryogels. The monolithic samples of cryogels used for column studies, after interaction with metal ions, were washed by 10 mL of 0.1 M HCl to remove adsorbed pollutants. After recovery, the column experiments were repeated with cryogel under the same conditions as described above. Five consecutive cycles of regeneration/adsorption were carried out. After each regener- ation, cryogels were washed with 50 mM NaOH and then UPW.

Figure 2. SEM, EDX elemental composition, and confocal microscopy images of the AAC (a, c, and e, respectively) and SAC (b, d, and f, respectively) cryogels.

3. RESULTS AND DISCUSSION

3.1. Synthesis and Characterization of Cryogels.The schematic representation of cryogels formation via cryopoly- merization technique is shown in Figure 1. To enhance the activity of allylamine and increase the yield of the gel fraction of copolymerized AMPS and AA, allylamine was converted to a phosphate complex by mixing with phosphoric acid.^50,51 Placing the monomeric solution at subzero temperatures initiates the growth of ice crystals, which occurs with radical polymerization of selected monomers, and the cross-linking of polymer chains to form branched macromers in unfrozen liquid microphase.⁵² The size of the formed ice crystals depends on the temperature in the cryobath. Polymerization goes on until free monomers run out, or until termination of all radicals occurs. Upon completion of the polymerization reaction and ice thawing, a hollow of microcrystals leads to macro-sized pores development within the polymeric structure.

The structural morphology and formation of macropores of the cryogels was observed by scanning electron microscope.

SEM images display a cross-linked polymeric network integrated with super-macropores with pore diameters of 10−50 μm (Figure 2a,b). While the general structure of cryogels is macroporous, the walls of polymers have slit- cylindrical micropores in a range of 5−20 nm (Figure 2c,d).

The images of the three-dimensional structure of the swollen cryogels were evaluated byfluorescence confocal microscopy at a depth of 150 μm, and processed with Imaris Software. It is clearly seen that the whole structure of the AAC polymer shows homogeneous distribution of macropores (Figure 2e).

The SAC cryogel has a more rigid and tight structure due to the presence of AMPS monomer (Figure 2f).

SEM/EDX analysis is often used for the determination of elemental composition of cryogels.^40,53The mass percentage of carbon and oxygen were found to be in approximately the same level for the AAC and SAC cryogels. Since the amount of NaOH used in AAC cryogel synthesis was higher than that in the SAC sample: The mass percentage of Na in thefinal AAC

was 9%, while in SAC it reached only 3.1%. In SAC cryogel, the main monomer was sulfur-containing AMPS, and the amount of sulfur in the adsorbent was equal to 20.9% (Figure 2a,b). The microporosities of the cryogel walls were also confirmed by the nitrogen adsorption/desorption porosimetry technique. The specific surface areas of cryogels were determined by a multipoint BET analysis, while the BJH and DR models were used to determine mesoporous and microporous surface areas, pore volumes, and pore diameters, respectively. The results of porosity determination are presented inTable 3. The BET surface area of AAC is 62.34 m² g⁻¹, a value about 60% smaller than the one obtained for SAC. Similarly, the specific micropore volume of AAC is almost half the one retrieved for SAC when applying the DR method. The results of surface area and pore volume measurements by BJH and DR methods are in line with SEM images (Figure 2c,d), where SAC, visually, has a higher amount of slit pores on the surface of polymer.

The presence of various functional groups in the cryogels was discussed in details in our previous papers.^40,52 Briefly, both AAC and SAC polymers show the existence of amide(I) and amide(II) groups due to the presence of pDMAAm and BisAAm reagents. Peaks of COO⁻carboxyl groups of amino acid residues and nondissociated carboxylic groups were determined at 1390 and 1141 cm⁻¹, respectively. Due to existence of AMPS monomer in SAC cryogel, the characteristic frequencies of sulfonic acid, sulfoxide, sulfide, and C−S functional groups were found. The zeta potential measure- ments results presented in our previous work showed that surface charge of AAC cryogel is negative in a wide range of pH, from 3.4 to 9, due to the deprotonation of carboxyl groups.⁵² A positive charge of AAC at pH below 3.2 is attributed to positively charged amino groups of allylamine. In the case of the SAC cryogel, the surface is negatively charged for the whole pH range due to the presence of permanently negative form of dissociated sulfonic acid, which is a strong acid. It can be also assumed that amino groups of allylamine Table 3. Microstructural Parameters of Cryogels

BET BJH DR

specific surface area [m²g⁻¹]

specific surface area [m²g⁻¹]

specific pore volume [cm³g⁻¹]

pore diameter

[nm]

specific micropore surface area [m²g⁻¹]

specific micropore volume [cm³g⁻¹]

pore diameter

[nm]

AAC 62.34 45.44 0.167 4.27 60.77 0.022 3.09

SAC 147.38 165.06 0.230 3.06 112.09 0.040 3.04

Figure 3.Equilibrium loading of metal ions (Sr²⁺and Cs⁺) by AAC (left) and SAC (right) cryogels at different pH values.

and neighboring sulfonic and carboxyl groups are protonated at low pH.⁵²

3.2. Effect of pH on Metal Ions Adsorption.The pH of the solutions after the dissolution of 100 ppm Sr²⁺ and Cs⁺ ions of nitrate salt was around 5.5. To avoid the precipitation of these metal ions (in the form of insoluble hydrocomplexes), solution pH values of ≤5.5 was chosen for the experiments.

Results of specific equilibrium adsorption capacity reported in Figure 3show that the removal of strontium and cesium ions monotonically decreases with the diminution of pH. At a highly acidic pH value of 1.0, the adsorption capacities of the AAC and SAC samples for Sr²⁺ and Cs⁺were around 44−47

and 33 mg g⁻¹, respectively. As the pH increased to 3.0, however, the removal efficiency doubled for Sr²⁺and Cs⁺ions, and the highest removal capacities of Sr²⁺and Cs⁺ions at pH 5.5 were 124 and 139 mg g⁻¹by AAC and 88 and 136 mg g⁻¹ by SAC, respectively. At low pH, H⁺ competes for ion- exchange sites in the cryogel structure, hindering the uptake of metals.⁵² Also, at higher pH values, excessive hydroxyl ions may cause a strong electrostatic attraction with the metal cations for the negatively charged adsorption sites, which causes sudden increases in adsorption capacity.

3.3. Adsorption Kinetics. The results of kinetics experi- ments are illustrated inFigure 4in terms of removal efficiency Figure 4.Removal efficiency of Sr²⁺(left) and Cs⁺(right) ions on the AAC and SAC cryogels (70 mg of cryogel in 100 mL of solution), at pH = 5.

Figure 5.Comparison between experimental and theoretical kinetics adsorption patterns for the investigated sorbent/adsorbate systems: (a) SAC- Cs⁺, (b) SAC-Sr²⁺, (c) AAC-Cs⁺, and (d) AAC-Sr²⁺.

as a function of process time. It is evident that the AAC sample shows a faster kinetics with respect to SAC sample in the removal of both cations in thefirst 8 h. In thefirst hour, AAC cryogel can remove about 42 and 45% of Sr²⁺ and Cs⁺, respectively, while SAC removed only about 14 and 21%, respectively. After 24 h of contact with the strontium- containing solution, both cryogels reached equilibrium with removal efficiency values of 86% for AAC and 48% for SAC sample. In the case of Cs⁺ions, both AAC and SAC cryogels also reached equilibrium in 24 h with maximum removal efficiencies of 58 and 63%, respectively.

The presence of adsorbed Sr²⁺ and Cs⁺ metal ions on the cryogels surface was verified by point EDX analysis and is presented in Figure S1. The overall picture of elemental analysis confirms that AAC cryogel has higher adsorption capability compared with SAC sample. The EDX analysis reveals that all adsorbed metal ions are evenly distributed over the entire surface of the polymers (see Figures S2−S5).

3.4. Kinetic Modeling Studies. The results of batch dynamic adsorption tests were modeled according to the theoretical framework described in section 2.4.2. Figure 5 reports a comparison between experimental and theoretical dynamic profiles of the adsorption capacity q t( ) for the investigated systems, while Table 4 summarizes the main parameters determined from the modeling campaign. As a general observation, there are not very remarkable differences in the fitting accuracy of the three models. From close inspection of dynamic adsorption profiles and based on theR² values, it can be said that the diffusional model accounting for the Chen and Yang’s expression for surface diffusivity (PVSDM-CY) exhibits slightly lessfitting accuracy and tends to anticipate the attainment of equilibrium conditions with respect to experimental observations, for all the sorbent/

adsorbate pairs. For both the SAC-Cs⁺ and SAC/Sr²⁺

adsorption systems, both the diffusional model with constant surface diffusivity (PVSDM) and the fractal-based one (PVSDM-MBI) produce practically the same kinetics patterns.

In the case of AAC cryogel and for both tested ions, the PVSDM-MBI shows a slightly superior ability in predicting q t( ) versus time trends with respect to PVSDM. For each adsorbent material, Cs⁺ exhibits faster sorption dynamics with respect to Sr²⁺, as evidenced by greater values of k_f, D_p, and D t_s( ) obtained for the former ion. This can be related to the greater value of molecular diffusivity of cesium with respect to strontium (seeD_minTable 1), which in turn is determined by a higher ionic mobility.⁴⁸When comparing adsorbent materials for the same cation, faster sorption dynamics were observed for AAC cryogel when compared to SAC. The mass transfer coefficient in thefluidfilm is almost 20 times greater for AAC- Cs⁺with respect to SAC-Cs⁺, while it is almost double for the AAC cryogel in the Sr²⁺case. Similarly, the adsorbate transport in the AAC pores network is faster than in the SAC case (cf.D_p values inTable 1), which can be related to the slightly greater macroporosity of the AAC adsorbent (cf.εpinTable 1and its favorable effects ineqs S4 and S5). The diffusion coefficients obtained for AAC and SAC cryogels was found in the range of 1.01 × 10⁻⁹−8.81 × 10⁻¹⁰ m² s⁻¹ and depends on the composition, cross-linking percentage, and ionic strength of the immersion solution. Taguchi et al.²⁸ synthesized resorcinol−formaldehyde hydrogel with various resorcinol (R) (main monomer) and sodium carbonate (C) (reaction catalyst) ratios for cesium uptake. It was found that, when the R/C ratio (mol/mol) was 2.5, the diffusion coefficient was 1.45×10⁻¹²m²s⁻¹, while with increase of the R/C ratio, the diffusion coefficient increased and reached 7.11×10⁻¹²m²s⁻¹. Scally et al.⁵⁴evaluated the diffusion coefficients of acrylamide monomer cross-linked with an agarose derivative (APA) hydrogel during Pb²⁺, Ni²⁺, Cd²⁺, and Cu²⁺ removal. At metal ion concentrations of 10 ppm, the diffusion coefficients were found to be 8.0×10⁻¹⁰m²s⁻¹for Pb²⁺, 5.77×10⁻¹⁰m² s⁻¹for Ni²⁺, 6.09×10⁻¹⁰m²s⁻¹for Cd²⁺, and 4.41×10⁻¹⁰m² s⁻¹ for Ni²⁺. Finally, the mean integral value of the fractal- based surface diffusivity is slightly greater for AAC with respect to SAC material. A deeper analysis of surface diffusivity Table 4. Main Parameters Determined from the Kinetic Modeling Campaign

SAC-Cs⁺ SAC-Sr²⁺ AAC-Cs⁺ AAC-Sr²⁺

PVSDM

k_f[m s⁻¹] 8.56×10⁻⁶ 5.81×10⁻⁶ 1.55×10⁻⁴ 1.08×10⁻⁵ D_s[m²s⁻¹] 5.00×10⁻¹¹ 2.85×10⁻¹¹ 1.76×10⁻¹⁰ 5.11×10⁻¹¹

R²[−] 0.9876 0.9650 0.9964 0.9187

PVSDM-CY

k_f[m s⁻¹] 8.56×10⁻⁶ 5.81×10⁻⁶ 1.55×10⁻⁴ 1.08×10⁻⁵ D_s0[m²s⁻¹] 5.00×10⁻¹¹ 2.85×10⁻¹¹ 1.76×10⁻¹⁰ 5.11×10⁻¹¹

λ[−] 0.50 0.94 0.01 0.42

R²[−] 0.9863 0.9639 0.9924 0.9188

PVSDM-MBI

k_f[m s⁻¹] 8.56×10⁻⁶ 5.81×10⁻⁶ 1.55×10⁻⁴ 1.08×10⁻⁵ D_s0[m²s⁻¹] 5.00×10⁻¹¹ 2.85×10⁻¹¹ 1.76×10⁻¹⁰ 5.11×10⁻¹¹

β[−] 0.25 0.12 0.24 0.11

h[−] 0.37 0.61 0.94 0.91

α[−] 0.20 0.20 0.84 0.47

[ ⁻]

D t_s( ) m s^{2 1} 4.26×10⁻¹¹ 2.29×10⁻¹¹ 5.43×10⁻¹¹ 2.87×10⁻¹¹

ρ [−]

D t q D C

s( ) p 0

p 0

1.58 0.79 2.55 1.74

Bi t( )[−] 13.98 17.83 166.92 19.66

R²[−] 0.9876 0.9651 0.9978 0.9204

parameters retrieved for the PVSDM-MBI reveals that for the same cation the fractional contribution of the fractal diffusion resistance (βparameter ineq 5) is similar for the two tested adsorbents. Moreover, theβvalue is more than 2-fold greater for Cs⁺ with respect to Sr²⁺ for the same adsorbent.

Additionally, the AAC sorbent displays greater values of both the hopping exponent α and the fractal exponent hfor each adsorbate; for example, h is 2.5 and 1.5 times greater in the AAC case with respect to the SAC case for Cs⁺ and Sr²⁺, respectively. These results are also mirrored by stronger time- dependence of the surface diffusivity ratio ^{D t}

D

s( )

s0

for AAC adsorbent as depicted inFigure 6. It can be observed that all

systems under investigation exhibit a nonmonotonic pattern of

D t D

s( )

s0

with a maximum reached within thefirst 2.5 h, and the

absolute values of the maxima are greater for AAC with respect to SAC cryogel. Moreover, the minimum values attained by the surface diffusivity ratio, i.e., under equilibrium conditions, are lower for AAC, which can be related to the greater values ofh determined for this sorbent (which in turn affects the time decrease of the surface diffusivity):^{D t}

D

s( )

s0

is 0.6 for SAC-Sr²⁺and 0.1 for AAC-Cs⁺ after 48 h. The analysis of the average Biot number shows that both external and internal resistances are important (with the internal one characterized by compara- tively higher relevance) for AAC-Sr²⁺, SAC-Cs⁺, and SAC/Sr²⁺

systems (Bi t( ) ranges from 14 to 20); however, Bi t( ) is 6 (10²) for AAC-Cs⁺, suggesting that in this case the intraparticle transport mechanism is significantly limiting with respect to thefluidfilm diffusion, linked to the remarkable greater value of k_f obtained for this system. Finally, a comparison of the values of surface diffusion and pores transport resistances through ^{D tρq}

D C

s( )p 0 p 0

values highlights that both intraparticle mechanisms are relevant (cf. also the discussion reported in Inglezakis et al.).⁴⁷

3.5. Effect of Metal Ions Initial Concentration and Cryogels Loading on the Adsorption Isotherms. To evaluate the equilibrium behavior of the cryogels, three equilibrium adsorption isotherms were studied by using different experimental methods. The results are presented in Figure 7a−d. In the first and third set of experiments (experiments 1 and 3, cf. Table 2), various cryogel weights ranging from 0.0022 to 0.9 g were used atfixed concentration (100 or 1000 mg L⁻¹); this is a typical method for studying ion-exchange isotherms which require constant liquid phase normality.⁵⁵The second set of isotherm experiments (experi- ment 2) was conducted under afixed cryogel mass of 0.07 g Figure 6.Surface diffusivity ratio as a function of time predicted by

the PVSDM-MBI for the investigated sorbent/adsorbate systems.

Figure 7.Isotherms of metal ions removal by cryogels: (a) AAC-Sr, (b) SAC-Sr, (c) AAC-Cs, and (d) SAC-Cs.

and 100 mL of solution, with variable initial concentration ranging from 100 to 1000 mg L⁻¹. This is the standard method used in adsorption studies.

The results show that the equilibrium solid-phase concen- tration increases with the increase of initial concentration of metal ions in the solution, as expected. The maximum adsorption capacity of the AAC cryogel was 362 mg g⁻¹ for Cs⁺ and 209 mg g⁻¹ for Sr²⁺. The SAC polymer performed better at Cs⁺ removal with respect to Sr²⁺, with maximum adsorption capacities of 259 and 211 mg g⁻¹, respectively. In the adsorption literature, the derivation of isotherms is commonly carried out by varying the aqueous phase concentration while keeping the adsorbent mass to solution volume ratio constant. Interestingly, there are no studies comparing the standard method with alternatives, for instance, by varying the sorbent mass to solution volume ratio under constant aqueous phase concentration of adsorbate. If only adsorption takes place and at constant temperature, then the different experimental approaches must result in a single isotherm. However, in many systems, a combination of mechanisms exists, and such a comparison of methods becomes interesting. For instance, it is well-known that the isotherms of ion-exchange systems depend on total normality, a phenomenon called concentration-valence effect.⁵⁵ Taking into account that the removal mechanism of metals on cryogels is complex, being a combination of reversible adsorption, complexation, and ion-exchange processes, the differences among the isotherms derived by using different experimental procedures should emerge, a phenomenon that requires further investigation and is out of the scope of this paper.

Being the most common experimental method in sorption studies, the results of experiment 2 were used for isotherms modeling with Langmuir and Freundlich equations. From the outcomes of the adopted models (Table S2), it was found that the correlation coefficient (R²) of the Freundlich isotherm model was higher than values obtained from Langmuirfitting for both cryogels in Sr²⁺solutions. In the case of Cs⁺ions, both models showed high values of R² with slightly better performance for Freundlich in the case of AAC cryogel.

These results indicate that both bivalent strontium and monovalent cesium ions can interact with various functional groups of polymers on the heterogeneous surface of the cryogels.

The maximum capacities achieved are comparable with those reported in the literature for 3D porous materials used

for the removal of Sr²⁺and Cs⁺from aqueous solutions. Garg et al.²⁰ synthesized a series of poly-N-vinyl imidazole-based hydrogels by adding four different cross-linkers for the removal of Sr²⁺ ions. The authors reveal that p(N-VIm-cl-DVB) hydrogel showed the best adsorption results with maximum capacity of 247 mg g⁻¹at pH 7.0. In another study, Choe and coauthors²¹ fabricated an alginate/humic acid/Fe-amino clay hydrogel for targeted removal of Sr²⁺ ions under different conditions of pH, temperature, and competing cations. The result of the study showed that in a single-metal solution the synthesized hydrogel was able to remove up to 46 mg g⁻¹of Sr²⁺ ions, while from multicomponent model solution, it was able to remove only 8.3 mg g⁻¹.²¹In case of Cs⁺removal, many studies were done by use of Prussian blue modified polymers and clays. Zhao et al.³³ reported the production of PAMPS/

PAAm hydrogel modified by [Fe(CN)₆], and examined polymers composites capability to removal Cs⁺ from 1000 ppm of CsCl solution at pH 7.0. The result showed the high uptake of cesium ions by coupling effect of Cs⁺ with the framework of Prussian blue crystals and reached the maximum capacity of 420 mg g⁻¹ of composite. In the studies of Bratskaya et al.,²⁷macroporous Co²⁺-chelated carboxyalkylchi- tosan cryogels were fabricated and modified with potassium ferrocyanide for selective capture of cesium ions. The results of the experiments showed that CEC/Co(II)[Fe(CN)₆] cryogel was able to adsorb up to 132 mg g⁻¹of Cs⁺ from 264 ppm solution at pH 3.0. More results on the removal of Sr²⁺and Cs⁺ from water by 3D polymeric adsorbents found in the literature are presented inTable S1.

3.6. Removal Mechanisms. The FT-IR analysis showed that some functional groups of cryogels can act as ion- exchangers and chelation agents toward cationic metals. The removal mechanisms proposed in our previous papers for Hg²⁺

and Cd²⁺ capture from aqueous solution may also be applied for Sr²⁺and Cs⁺ ions removal.^42,52The mass of Na⁺ released from cryogels was measured, and the Me/Na⁺(with Me = Sr²⁺

and Cs⁺) molar ratio is shown inFigure 8as a function of the mass of cryogel perfixed volume of metal ion solution. Ideally, if the ion exchange of Sr²⁺ occurs only with Na⁺, then the molar ratio is 1:2, and for Cs⁺, it is 1:1. In the case of AAC, the ratios are not far from the expected ones for ion exchange, and the deviations can be explained by the contribution of the functional groups. However, in the case of the SAC cryogel, for both target adsorbate ions, the Me/Na⁺ ratio is much higher than the expected one for ion exchange, which means that the Figure 8.Molar ratio of removed metal ions to released Na⁺during adsorption as a function of the sorbent mass (metal ion solution volume of 100 mL).

released amount of Na⁺is small. This happens due to the low initial concentration of sodium ions in the SAC cryogel (see section 3.1) and probably also because of the sulfonic acid’s H⁺, which might be exchanged for Sr²⁺and Cs⁺ cations.

The results of XPS characterization of cryogels before and after adsorption of the metal ions (Figure 9a,b) show that peaks of sodium ions Na 1s (∼1071.4 eV) and Na Auger peak (∼497.1 eV) of the parent AAC and SAC cryogels disappeared with the advent of corresponding metal ions peaks of strontium at 135.1 eV and cesium at 725.1 and 738.1 eV, indicating the relevance of the ion-exchange mechanism of sodium ions to selected cations. After the interaction of both cryogels with the studied cations, the binding energies of C−O

(286.8 eV) and CO (288.2 eV) are shifted by 0.7−1.2 eV to higher positions, due to the binding of positively charged metal ions.⁵⁶Since most of the components for SAC are the same as those for AAC, the main difference is the presence of sulfonic acid groups in the structure of SAC cryogel.Figure 9b shows spectra of SAC cryogel before and after metal ions removal.

The S 2p core level spectrum is deconvoluted into two major peaks centered at 167.5 and 169.0 eV, which are assigned to 2p_3/2 and 2p_1/2. These high oxidized states of sulfur are corresponding to sulfonic acid group (−SO³⁻).⁵⁷ After interaction of metal ions, the sulfonic acid group peaks shifted of 0.5−0.7 eV to higher region, which suggests the formation of coordination bonds of sulfonic acid residues with cations.

Figure 9.XPS patterns of cryogels before and after metal ions removal: (a) AAC and (b) SAC cryogels.

These findings indicate a combination of ion-exchange, adsorption, and chelation with the involvement of the carboxylic and sulfonic groups, resulting in the high adsorption capacity of the synthesized cryogels.

3.7. Leaching Experiments.AAC and SAC cryogels were immersed in water for 30 days (neutral pH) after Sr²⁺and Cs⁺ adsorption, in order to verify undesired leaching phenomena.

Results reported inTable 5showed negligible leaching of Sr²⁺

metal ions, with values smaller than 1% for both cryogels. In the case of Cs⁺ ions, leaching was greater than that for strontium and reached 2.30 and 1.84% for AAC and SAC, respectively. Presumably, this phenomenon is associated with the monovalence of cesium and a weaker binding with functional groups of cryogels with respect to the strontium case.

3.8. Removal from Various Water Matrices and Comparison to Commercial Adsorbents. The results of Sr²⁺ and Cs⁺ removal from various water matrices and commercial adsorbents are presented in Table S3. The concentrations of coexisting cations in tap water were 77.3 ppm Na⁺, 5.4 ppm K⁺, 20.5 ppm Mg²⁺, and 61.3 ppm Ca²⁺, while in river water they were 19.7 ppm Na⁺, 1.7 ppm K⁺, 3 ppm Mg²⁺, and 29 ppm Ca²⁺.

In the case of Cs⁺removal from UPW, in thefirst 30 min of experiment the zeolite and AAC sample removed around 53 and 37%, respectively, while the SAC adsorbed around 27%, and the ion-exchange resin removed only 5%. After 24 h of processing time, 86−87% of cesium ions were removed by commercial zeolite and AAC cryogel, and 60 and 39% were removed by SAC and ion-exchange resin, respectively. In tap and river water, due to the occurrence of coexisting cations, the

removal effectiveness of all materials was lowered. However, the general pattern of removal was the same as that for UPW.

After 24 h, the AAC cryogel and commercial zeolite were able to remove around 40 and 50% of Cs⁺ from tap water and 54 and 55% of Cs⁺ from river water, respectively, whereas the SAC sample and ion-exchange resin did not show any adsorption of cesium from tap water and only 21 and 27%

from river water after 24 h of interaction, respectively.

As in the case of Cs, for the removal of Sr, commercial zeolite performs better in thefirst 30 min in all water matrices;

however, the removal of strontium from UPW was higher and reached 66% in thefirst 30 min and more than 97% after 24 h.

After 24 h in UPW, the AAC cryogel reached 95% removal of Sr²⁺, while the SAC and the ion-exchange resin reached 80 and 36% removal. The tap water results showed that only SAC cryogel exhibited low effectiveness (around 27%), while the other ion-exchange materials showed 41−45% Sr²⁺ removal.

The results of Sr²⁺ removal from river water followed the following ranking: commercial zeolite (64%) >AAC (50%) >

SAC (49%) ≈ion-exchange resin (49%).

3.9. Dynamic Adsorption Experiments. The cryogels were applied in afixed-bed column for dynamic sorption, and results are reported in Figure 10. Data show that the breakthrough curve in the case of Cs is shifted toward longer times in comparison to Sr; for example, in the case of AAC cryogel, saturation occurs after about 35 and 20 min for Cs⁺ and Sr²⁺, respectively. For the same cation, AAC displays better dynamic performance when compared to SAC cryogel. These results are in agreement with the kinetics (surface diffusivity) and equilibrium (capacity) results.

The breakthrough curves for the removal of the cryogels were modeled using the Thomas, Adams−Bohart, and Yoon− Nelson models, by applying linear regression analysis to determine the dynamic behaviors in the column. Comparisons of predicted model parameters together with values of experimental and theoretical adsorption capacity are shown inTable S4. Results show that that correlation coefficients for the Thomas and Yoon−Nelson models are practically equivalent and higher than values retrieved for the Adams− Bohart model. As an example, k_Th values are practically equivalent for AAC and SAC when considering Sr²⁺

adsorption, and at fixed adsorbent levels, the rate constant Table 5. Leaching Results of Metal Ions from AAC and SAC

Cryogels at pH 6.5^a

adsorbent

total adsorbed metal ions [mg]

leached metal ion after 30 days [mg]

leached metal ion after 30 days [%]

AAC-Sr 8.89 0.0656 0.74

AAC-Cs 6.88 0.1585 2.30

SAC-Sr 4.45 0.0418 0.94

SAC-Cs 5.73 0.1054 1.84

aExperimental conditions: 0.07 g of sorbent in 100 mL of 100 mg L⁻¹ metal ions solution.

Figure 10.Breakthrough curves for AAC and SAC cryogels (adsorption of Cs⁺and Sr²⁺).

values for strontium are slightly greater than figures retrieved for cesium.

3.10. Regeneration and Reusability of Cryogels. The recovery and reuse of the adsorbent for multiple cycles is one of the most important aspects from an economic point for industrial scale processes. Cryogel monoliths used in the column experiments were washed with 0.1 M HNO₃and then by 0.1 NaOH and UPW. The removal efficiencies of cryogels toward Sr²⁺and Cs⁺duringfive cycles are presented inFigure 11. After the first cycle of desorption/adsorption, the efficiencies of both cryogels were almost 100%. During the next cycles of regeneration and reuse, the adsorption behavior started to decrease and was found to be 96% for AAC-Sr, 97%

for AAC-Cs, and 92 and 95% for Sr²⁺and Cs⁺removed by the SAC cryogel, respectively. The removal efficiency decreased by 8% after five cycles, thus evidencing that the effectiveness of both cryogels was still high.

4. CONCLUSIONS

Two types of macroporous cryogels were studied toward Sr²⁺

and Cs⁺ removal from aqueous solutions. The results demonstrated fast removal kinetics and high equilibrium capacities with the AAC cryogel being superior to the SAC cryogel. In particular, the adsorption capacity of the AAC cryogel reached 362 mg g⁻¹for Cs⁺and 209 mg g⁻¹for Sr²⁺. In contrast, the SAC polymer was also better at Cs⁺removal, with a maximum capacity of 259 mg g⁻¹, while Sr²⁺ adsorption reached 211 mg g⁻¹. The combination of experimental, modeling, and characterization indicates that ion exchange followed by complexation reactions is the main mechanism of cations removal. Kinetics modeling by use of advanced diffusion-based equations showed that both external and internal resistances are important, and surface diffusivity displays a nonmonotonic pattern as predicted by a fractal- based expression. Column experiments demonstrated that the sorbents still retained more than 90% of their capacity afterfive adsorption/desorption cycles, which makes the cryogels very promising candidates for the scale-up of continuous adsorption units aimed at the removal of strontium and cesium from wastewaters.

■

ASSOCIATED CONTENT

*^sı Supporting Information

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.iecr.2c00531.

Adsorbents used in the literature for the removal of Cs⁺ and Sr²⁺ ions from aqueous media, isotherms modeling results, residual concentrations of Cs⁺and Sr²⁺in various water matrices after adsorption, breakthrough models parameters for adsorption of Cs⁺and Sr²⁺onto cryogels, SEM/EDX mapping images of cryogels after adsorption of Sr²⁺ and Cs⁺ metal ions, mathematical modeling for batch and column adsorption (PDF)

■

AUTHOR INFORMATION Corresponding Authors

Vassilis J. Inglezakis −Department of Chemical&Process Engineering, University of Strathclyde, Glasgow G1 1XJ, United Kingdom; Email:vasileios.inglezakis@strath.ac.uk Marco Balsamo −Department of Chemical Sciences,

University of Naples Federico II, Complesso Universitario di Monte Sant’Angelo, 80126 Napoli, Italy; orcid.org/0000- 0002-2063-8680; Email:marco.balsamo@unina.it Authors

Alzhan Baimenov−National Laboratory Astana, Nazarbayev University, Nur-Sultan 010000, Kazakhstan;

Faculty of Chemistry and Chemical Technology, Al-Farabi Kazakh National University, Almaty 050012, Kazakhstan Fabio Montagnaro−Department of Chemical Sciences,

University of Naples Federico II, Complesso Universitario di Monte Sant’Angelo, 80126 Napoli, Italy; orcid.org/0000- 0002-6377-3989

Complete contact information is available at:

https://pubs.acs.org/10.1021/acs.iecr.2c00531

Notes

The authors declare no competingfinancial interest.

■

NOMENCLATURE

Bi t( ) average Biot number for PVSDM-MBI [−] C(t) adsorbate concentration at thefixed bed outlet at

timet[mg m⁻³]

Figure 11.Removal efficiency of AAC and SAC cryogels for adsorption of Cs⁺and Sr²⁺duringfive adsorption/desorption cycles.